The Annual Business Inquiry (ABI) is conducted in two parts: one dealing with employment, the other with financial information. The financial inquiry covers about two thirds of the UK economy i.e. production; construction; motor trades; wholesale; retail; catering and allied trades; property; service trades; agriculture (part), hunting, forestry and fishing. The coverage of the employment inquiry is wider.
The ABI is carried out by the Office for National Statistics. UK Businesses are sampled according to their employment size and industry sector.
Results are used to compile the input-output tables in the National Accounts, rebase the Index of Production, benchmark the inventories and capital expenditure quarterly inquiries and are supplied to Eurostat under the European Structural Business Statistics Regulation.
Data are available, on the website at www.statistics.gov.uk/abi, for 13 main variables down to the 4-digit level of SIC(92)/ SIC(2003) and includes higher aggregates. These variables include Total Turnover, Approximate Gross Value Added, Total Purchases, Employment Costs and Capital Expenditure.
Regional data is available for Government Office Regions, Scotland, Wales and Northern Ireland at the 2 digit Division level of SIC(92)/SIC(2003).
Data is also available for Retail Commodities at the 3 digit Group level of SIC(92)/ SIC(2003) and Retail Outlets data at the 4 digit Class level of SIC(92)/SIC(2003).
Data not available on the website may be available either as a standard extract or as a special analysis, which may or may not be chargeable.
Information on these services can be obtained from the ABI Analysis Page
The website replaces the existing ABI publications: Production and Construction Inquiries Summary Volume (PA1002); PACSTAT CD Rom and Distribution and Services Sector Reviews which will no longer be published.
The 2007 inquiry selected approximately 66,600 businesses
Linked surveys/sources:
ABI 1, Quarterly Capital Expenditure, MPI, MIDSS, PRODCOM, RSI, ITIS, R&D, E-Commerce, Quarterly Stocks
Method:
Other large-scale (Sample) Survey
Status:
Ongoing
Frequency of collection or compilation:
Annually
Reference period:
Calendar year but business year data accepted
Timeliness:
Provisional results are available 11 months after reference year. Publication of final results available 18 months after reference year.
Year data first available:
1998 for ABI data (including Regional), 1970 Production and Construction Industries, 1976 Distribution and Service Industries
Year of latest available data:
2007
History of data collection / breaks and discontinuities:
The ABI was set up in 1998 and is the combination of the Annual Census of Production & Construction (ACOP/ACOC) and Distribution and Services Inquiries.
The Annual Census of Production began in 1907 and continued at roughly five-yearly intervals until it became annual in 1970. Census information for this year onwards is available in the Annual Summary Volume up to and including 1997.
The Distribution and Services Inquiries began in 1976 and were published in summary volumes up to and including 1997.
Main areas for which data is collected:
National: UK Great Britain England Scotland Wales Northern Ireland
Extent of geographical coverage:
Sampled
Spatial units of data collection:
All registered businesses (reporting units, which are usually equivalent to enterprises) listed in required Sections of the SIC(92)/ SIC(2003) on the Inter Departmental Business Register.
Smallest spatial unit for which data are made available:
Available at Government Office Region (GOR), County, District, Nomenclature of units for Terrotorial Statistics (NUTS), Ward and Unitary Authority level.
Commonly available units for which aggregate data are made available:
Data is available to NUTS 4 (districts/unitary authorities) level on request. A charge may be made.
Standard geographical classification or coding systems used:
The geographic referencing system used is Government Office Regions (GOR) and Nomenclature of Units for Territorial Statistics (NUTS)
Descriptive summary of geographical coverage and geographic referencing system:
Survey covers UK, Great Britain, England, Scotland, Wales and Northern Ireland
The geographic referencing system used is Government Office Regions (GOR) and Nomenclature of Units for Territorial Statistics (NUTS)
Legislative status:
Statutory using the Statistics of Trade Act 1947 and the Statistics of Trade and Employment (Northern Ireland) Order 1988
Deposited with data archive?
Bibliographic material:
A web site copy of the Economic Trends Article November 2000 is available on the web site Report on the Quinquennial Review May 2004
Annual Business Inquiry - Sectors A - O of the Standard Industrial Classification (Revised 1992)/ SIC(2003) with some exceptions. Data collected on turnover, employment costs, purchases, taxes paid, capital expenditure, stocks and gross value added.
Total Turnover Approximate Gross Value Added Purchases of goods and services Total Stocks(end of year) Employment TotalTurnover Total Stocks(beginning year) Total capital expenditure(acquisitions & disposals)
Collection is via paper questionnaire. The majority of respondents receive a standard form type but derivations of this standard form type exist to cover peculiarities of certain industries. For most of these full form types there is a corresponding short form type, which does not contain the detailed breakdowns, requested on the full forms. A proportion of respondents in each sizeband will receive the short form in order to minimise the burden on businesses. The proportion varies by size so that most small businesses receive the short form while all large businesses receive the full form.
Main historical changes in methodolgy and other breaks/discontinuities:
From the publication of revised 2002 and 2003 data on 23 June 2005 Northern Ireland data has been processed by ONS.
For 2002 the Department of Enterprise Trade and Investment in Northern Ireland (DETINI) have conducted their own Annual Business Inquiry. This has meant there are methodological differences between the way the results for Northern Ireland and the regions within Great Britain (GB) have been calculated for the UK regional results. This means the results produced by ONS for Northern Ireland for 2002 are not consistent with those produced for 2001 when the ABI was a United Kingdom survey. For the 2003 inquiry period there was a change introduced into ONS ABI procedures to ensure Northern Ireland data was processed on a parallel basis to GB. However it was not possible for ONS to make this change within its systems for 2002 data and this may have led to some discontinuity in the Northern Ireland estimates between the two years. In line with ONS revisions policy, data for 2002 will not now be amended. However, as a one-off exercise, ONS has in conjunction with DETINI examined the growth in Gross Value Added (GVA) between 2002 and 2003. That exercise suggests that if the two ONS inquiries had been conducted on a similar basis, the Northern Ireland GVA for the industries covered by the ABI would have been estimated to have increased within the range of 7 to 9 percent.
The results of the 1998 survey have been produced using the new methodology of the ABI that has brought together the annual inquiries covering production, construction, distribution and service sectors, standardising them into one integrated inquiry. This new system included a number of changes in methodology (outliers, estimation etc)
The Annual Business Inquiry for production and construction was previously known as the Annual Census of Production. A full census has not been conducted since 1968; thereafter sample surveys were introduced. Around 20,000 businesses were selected covering Production and Construction, Sections C-F of the Standard Industrial Classification Revised 1992 (SIC(92)). From 1994 stratified random sampling was introduced. The strata used are industry and employment size group. The sampling fractions vary by industry, but in general smaller businesses have a lower chance of selection. All the larger businesses in each industry are selected.
The Annual Business Inquiry for the distribution and services sector (ABI) represents an amalgamation of the annual inquiries into service trades, catering, property, wholesaling, motor trades and retailing. In recent years the range of variables collected has been increased. In 1992 the regular introduction of questions on detailed purchases and on employment costs began. In 1993 the coverage of the inquiries was extended to include Northern Ireland. 1995 data was based on employment sampling for the first time, prior to this sampling was based on turnover.
Service Trades - Since 1996 the coverage of the inquiry has been extended. The extensions include rail, sea and air travel, the national postal sector and membership organisations.
Catering and Allied Trades - Between 1960 and 1979 there was a benchmark inquiry into catering roughly every four years or so, but from 1979 the inquiry became annual.
Property - There has been a property inquiry since the mid 1950s but until 1994 data was only collected on capital expenditure. From 1995, the range of data was extended to bring the inquiry in line with the other DS inquiries.
Wholesaling and Dealing - The first major inquiry into wholesaling and dealing was carried out in respect of 1950, as part of the Census of Distribution. Subsequently, periodic large-scale detailed inquiries were conducted in respect of 1959, 1965, 1974 and 1990, but simpler annual inquiries were conducted for most intervening years and for all years since 1991.
Motor Trades - The first major inquiry into motor trades was carried out in 1950 as part of the Census of Distribution. Subsequently, periodic large-scale inquiries were conducted in respect of 1962, 1967 and 1972 although simple annual inquiries were carried out in most intervening years. By 1977 the annual inquiry was collecting detailed information on turnover and purchases.
Retail - From 1950 periodic Censuses of Distribution were conducted, the last of which was in 1971. Full-scale inquiries covering every retail business and every retail outlet were taken for 1950, 1961 and 1971, with large-scale inquiries for 1957 and 1966. The first annual retailing inquiry was conducted in respect of 1976 with a sample of 30,000 units. Throughout the late 1970s and '80s the inquiry varied from year to year in terms of both sample size and the amount of information collected. From 1991-1997 the sample remained reasonable constant at around 12,000.
The Inter Departmental Business Register (IDBR) is a register of businesses, the majority of which are registered for VAT and/or PAYE in the UK. It has recently been expanded to include businesses with low turnovers that did not appear on the register for VAT. The IDBR provides the basis for selection with the new levels being used in the 1996 ABI. Back data was reworked to bring it up to new levels.
Statistical procedures applied to data:
Data validation Imputation process Outlier detection Estimation
Details of validation and editing:
A wide variety of checks are undertaken on the data, including checks for missing key variables, internal consistency and credibility checking against the previous years data or industry specific validation gates. Where short period survey data exists for variables such as turnover, employment, capital expenditure and inventories, these data are checked for congruency. Between 25-30 different validation and consistency checks take place on a standard form type.
Imputation process:
Imputation is carried out for large non-responding businesses. Two medians are used in the ABI imputation process, 'link' and 'register'. A link median is calculated by comparing good returns for the current and previous year where the return is >0 and contributors employment is >99. There must be 3 or more matched pairs for this median to be used. A register median is calculated by comparing good returns for the current year with their register selected turnover (register employment for employment questions) where the contributors register employment is >99.
Outlier detection and treatment:
Outliers are detected based on a calculation of the ratio of returned to selected employment and returned to selected turnover. These extreme contributors are excluded from grossing and carry their returned values. Manual outliers are identified by the results sections, and are treated in the same manner in results as automatic outliers.
Estimation method:
Ratio estimation is used to produce UK estimates from the sample information. Estimates are produced for each industry by employment sizeband stratum and the stratum estimates are aggregated to produce population estimates. Register turnover is used as the auxiliary variable for the majority of variables but the registered employment is used as the auxiliary variable for wages and salaries.
Regional Estimation The local unit estimation process calculates an estimation factor for non-response from the apportioned local unit data. The estimation factor is then applied to the apportioned local unit values to give a total estimate for the local unit population. The estimation factor is calculated using the IDBR employment, classification and region of the local unit. This estimation factor will cause the regional data to differ from the estimated national data and so the data is constrained to the UK all industry level.
Seasonal adjustment or trend estimation technique:
N/A
Status of data:
Provisional results are available 11 months after the year. Final results are published 18 months after the year.
All businesses in Sections A - O, excluding J and L and parts of A and N, of the Standard Industrial Classification Revised 1992 (SIC(92))/SIC(2003).
Study population:
N/A
Sampling unit:
All registered businesses (reporting units, which are usually equivalent to enterprises) listed in required Sections of the SIC(92) /SIC(2003) on the Inter Departmental Business Register.
Sampling frame:
The frame is the Inter Departmental Business Register (IDBR). The main sources of information held on the IDBR are administrative details of Value Added Tax (VAT) traders supplied by HM Revenue and Customs (HMRC). These are supplemented by information on company links from independent sources, company registrations and feedback from ONS statistical survey.
Design/selection of sample:
The sample design is a stratified random sample. Stratification variables are employment sizeband, SIC(92)/SIC(2003) industry and countries (England and Wales, and Scotland). The basis for the sample allocation is the Neyman method. Northern Ireland data is collected separately by DETINI.
Quality Measurement:
Sampling errors are available from 18th September 2002.
Sampling error:
Estimates of sampling errors are produced for 4 main variables; Total Turnover, Approximate Gross Value Added, Total Purchases and Total Net Capital Expenditure at 3 digit Group level of SIC(92)/SIC(2003) using standard formulae.
Further information can be found on the Quality Measures page at www.statistics.gov.uk/abi
Non-sampling error:
The response rate gives an indication of the likely impact of non-response error on the survey estimates. Response rates are calculated by section and can be found on the Quality Measures page at www.statistics.gov.uk/abi
Restrictions on access/usage:
Legislation used to conduct the survey (Statistics of Trade Act, 1947) protects suppliers of data by making disclosure of information about individual units a criminal offence. ONS use a range of tests for disclosure and appoints a Data Custodian who is responsible for maintaining confidentiality. No information is published or released unless it is satisfied that the risk of disclosure of information about individuals and subjects, even indirectly, is effectively eliminated. Permission to publish potentially disclosive data may be sought from individual units.
Method of selection and linkage:
Reporting unit ; permanent random number. Continual selection in the 250+ employment sizeband. Osmotherly rotation in the 0-9 employment sizeband. Standard rotation in all other sizebands.
Size:
The 2006 inquiry selected approximately 66,400 businesses
ABI Data